Combining Monte Carlo and Molecular Dynamics Simulations for Enhanced Binding Free Energy Estimation through Markov State Models

• DOI: 10.1021/acs.jcim.0c00406
• Publication Date: 2020-07-09T16:03:42Z
• Authors: Joan F. Gilabert; Oriol Gracia Carmona; Anders Hogner; Victor Guallar
• Keywords: Umbrella sampling; Thermodynamic integration; Rare events; Energy landscape; Parallel tempering

We present a multistep protocol, combining Monte Carlo and molecular dynamics simulations, for the estimation of absolute binding free energies, one most significant challenges in computer-aided drug design. The protocol is based on an initial short enhanced simulation, followed by clustering ligand positions, which serve to identify relevant states unbinding process. From these states, extensive simulations are run estimate equilibrium probability distribution obtained with Markov State Models, subsequently used energy. tested procedure two different protein systems, Plasminogen kringle domain 1 Urokinase, each multiple ligands, aggregated length 760 μs. Our results indicate that sampling events largely facilitates convergence subsequent exploration. Moreover, capable properly rank set ligands examined, albeit computational cost the, more realistic, Urokinase complexes. Overall, this work demonstrates usefulness methods regular simulation techniques as way obtain reliable affinity estimates.